Unraid v7 beta 3 and 4 Broken nfs server was fine in beta2

Restarting the service fixes the issue

 /etc/rc.d/rc.nfsd restart

root@BMM-Unraid:/etc# /etc/rc.d/rc.nfsd status
NFS server daemon is currently running.
root@BMM-Unraid:/etc# /etc/rc.d/rc.nfsd restart
rc.nfsd: Restarting NFS server daemon...
rc.nfsd: Stopping NFS server daemon...
rc.nfsd: NFS server daemon...  Stopped.
rc.nfsd: Starting NFS server daemon...
rc.nfsd: NFS server daemon...  Started.
root@BMM-Unraid:/etc# /etc/rc.d/rc.nfsd status
NFS server daemon is currently running.
root@BMM-Unraid:/etc# showmount -e localhost
Export list for localhost:
/mnt/user/bitcoin 192.168.2.251
/mnt/user/Backups 192.168.2.252
root@BMM-Unraid:/etc# 
 

Restarting the rc.nfsd service only will bring nfs server data bak up. However rpc still breaks and nfs doesn't work as it should.

Using unraid script from documentation:
https://unraid.net/de/blog/deploying-an-unraid-nfs-server

from setup nfs static ports

#Unriad change rpc ports:
## Configuration Variables
#DEFAULT_RPC="/etc/default/rpc"
#STATD_PORT=32766
#LOCKD_PORT=32768
#MOUNTD_PORT=32767
#RC_NFSD="/etc/rc.d/rc.nfsd"
#
## Function to configure NFS ports
#nfs_config() {
#    set -euo pipefail
#    
#    # Update the default RPC configuration file with the specified ports
#    sed -i \
#        -e 's/^#RPC_STATD_PORT=.*/RPC_STATD_PORT='"$STATD_PORT"'/' \
#        -e 's/^#LOCKD_TCP_PORT=.*/LOCKD_TCP_PORT='"$LOCKD_PORT"'/' \
#        -e 's/^#LOCKD_UDP_PORT=.*/LOCKD_UDP_PORT='"$LOCKD_PORT"'/' \
#        "$DEFAULT_RPC"
#
#    # Update rc.nfsd to specify the mountd port
#    sed -i 's/^\s\{4\}\/usr\/sbin\/rpc\.mountd$/    \/usr\/sbin\/rpc.mountd -p '"$MOUNTD_PORT"'/' "$RC_NFSD"
#
#    # Restart NFS services
#    /etc/rc.d/rc.rpc restart
#    sleep 1
#    /etc/rc.d/rc.nfsd restart
#}
#
## Execute the configuration function
#nfs_config
#
## Check if the configuration was successful
#if [[ $? -ne 0 ]]; then
#    /usr/local/emhttp/webGui/scripts/notify -i warning -s "NFS config failed"
#fi
#

Helps but doesn't fix the issue. 

other commands

shomemount -e %unraid IP%

and 

exportfs -v

 

show content after a service restart only public shares seem to work. can't get private not secure share settign to function from web ui. 

Even though 

exportfs -v shows rw nfs share is read only...

I don't know why nfs is doing weird things for me, i have tested on borg system with simlar issues.

I ended up having to run this user script to fix the issues

#!/bin/bash
sleep 10

/etc/rc.d/rc.nfsd stop
/etc/rc.d/rc.rpc stop

/etc/rc.d/rc.rpc start
/etc/rc.d/rc.nfsd start

# Restart NFS services
#/etc/rc.d/rc.rpc restart
#sleep 1
#/etc/rc.d/rc.nfsd restart

Stopping the service at first array start and then starting the service seems to fix the issues with nfs...
 

Somewhat fixed in beta 4.

I still have to stop the server and start them,

If unraid doesn't have any nfs share before the update and after beta4 udpate there is a RPC error.
If you edit a share setting and eanble a NFS share nfs doesn't restart nor enable nfs.

When running showmount -e i get rpc error no inatlized.

once a nfs sahre in this case to test public.
i had to reboot unraid and stop and start the rpc/nfs service to get nfs working again.

Some weridness with nfs still goin on... This wasn't a issues in beta 2...

borg system log:

 

 

...

Edited Tuesday at 08:35 PM by bmartino1

after a nfs share is created and unraid rebooted.

running the script to sleep stop rpc nfs and start rpc nfs fixes teh issue..

Attached is the borgs frist boot upgrade to beta4 of betmetal test machine borg before restart and after....
Looks to be a issue in RPC again...

?Service level order. rpc needs to start before nfs...

same on bmm which is a vm in proxmox... had issues but a server stop start fixed it...

Again, this wasn't a problem on beta2...

NFS needs looked at from beta 2 before issues were presented...

the-borg-diagnostics-20241022-1535.zip the-borg-diagnostics-20241022-1546- after reboot.zip bmm-unraid-diagnostics-20241022-1550.zip

Edited Tuesday at 08:52 PM by bmartino1

What is this script doing?  I see NFS mentioned in the name.
 

Oct 22 15:20:32 The-Borg emhttpd: /usr/local/emhttp/plugins/user.scripts/backgroundScript.sh "/tmp/user.scripts/tmpScripts/Beta3_Bringback_mc/script" >/dev/null 2>&1/usr/local/emhttp/plugins/user.scripts/backgroundScript.sh "/tmp/user.scripts/tmpScripts/Fix-Hosts/script" >/dev/null 2>&1/usr/local/emhttp/plugins/user.scripts/backgroundScript.sh "/tmp/user.scripts/tmpScripts/FixShareServers-NFS_and_Samba/script" >/dev/null 2>&1

 

Which one ar you referring too, there are quite a few in that line...

bring back mc add a mc_cd if then to /etc/profile as noted in another post:

Fix host adds the hostname to the host file due to unfi dns issues and fix common problems.
Simlar to the smb script I made, I have a host file in the config folder. I replace the host file.

script:

#!/bin/bash
sleep 5
rm /etc/hosts
cp /boot/config/hosts /etc/hosts
chmod 644 /etc/hosts

as example:
This is added to the host file: 
#Unraid needed:
127.0.0.1 BMM-Unraid.localdomain
127.0.0.1 BMM-Unraid
#Other Docker Services
192.168.2.254 BMM-Unraid

to fix the dns error in FCP and dns for host ip name...
I disable this script, cat if any changes and update host based on other dns needs...

Fix share services is the scritpt above and in pictures to stop nfs, rpc and start...

 

On 10/23/2024 at 10:56 AM, dlandon said:

What is this script doing?  I see NFS mentioned in the name.
 

Oct 22 15:20:32 The-Borg emhttpd: /usr/local/emhttp/plugins/user.scripts/backgroundScript.sh "/tmp/user.scripts/tmpScripts/Beta3_Bringback_mc/script" >/dev/null 2>&1/usr/local/emhttp/plugins/user.scripts/backgroundScript.sh "/tmp/user.scripts/tmpScripts/Fix-Hosts/script" >/dev/null 2>&1/usr/local/emhttp/plugins/user.scripts/backgroundScript.sh "/tmp/user.scripts/tmpScripts/FixShareServers-NFS_and_Samba/script" >/dev/null 2>&1

 

 

Try 7.0 beta 4.4 and don't run your NFS stop/restart script.  There have been issues with stopping nfs in Linux in the last few Unraid releases and some Unraid changes have been made that may help your situation.

So i will ahve to asume that a typo and you meant 4 not 4.4 

same issues rpc not initalied..

ok will try: to upgrade....

will disable all user scripts:




? I will when its released to test.  I don't see unraid v7.0 beta 4.4

Edited yesterday at 12:46 AM by bmartino1

6 hours ago, bmartino1 said:

So i will ahve to asume that a typo and you meant 4 not 4.4 

same issues rpc not initalied..

ok will try: to upgrade....

will disable all user scripts:




? I will when its released to test.  I don't see unraid v7.0 beta 4.4

4.4 is an internal beta it I'll be in 5.

Thank you for that information. As I still need to apply my custom smb.conf. I have found that if i wait 1 min after boot so sleep 60. My samba smb works fine and as i want it too, but anything done to NFS requires UnRaid to be rebooted to take affect.

I have tested reboots, edits, no conf, a public nfs, no smb edits, no scripts at start... I keep coming back to something broken in rpc.

While I agree to a static port for RPC/NFS:
https://unraid.net/de/blog/deploying-an-unraid-nfs-server

ATM this breaks or makes things worse in beta 4.

atm i at least have a working work around. although, I don't like the idea that I have to reboot unraid each time I make and add a nfs share.